Direct evidence for occlusion in stereo and motion

  • James J. Little
  • Walter E. Gillett
Stereo And Motion
Part of the Lecture Notes in Computer Science book series (LNCS, volume 427)


Discontinuities of surface properties are the most important locations in a scene; they are crucial for segmentation because they often coincide with object boundaries. Standard approaches to discontinuity detection decouple detection of disparity discontinuities from disparity computation. We have developed techniques for locating disparity discontinuities using information internal to the stereo algorithm of [2], rather than by post-processing the stereo data. The algorithm determines displacements by maximizing the sum, at overlapping small regions, of local comparisons. The detection methods are motivated by analysis of the geometry of matching and occlusion and the fact that detection is not just a pointwise decision. Our methods can be used in combination to produce robust performance. This research is part of a project to build a “Vision Machine” [7] at MIT that integrates outputs from early vision modules. Our techniques have been extensively tested on real images.1


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. [1]
    M. Bertero, T. Poggio, and V. Torre. Ill-posed problems in early vision. Proc. of the IEEE, 76(8):869–889, Aug. 1988.Google Scholar
  2. [2]
    M. Drumheller and T. Poggio. On parallel stereo. In Proc. of IEEE Conf. on Rob. and Auto., pages 1439–1448, Washington, DC, 1986. IEEE.Google Scholar
  3. [3]
    W. Gillett. Issues in parallel stereo matching. Master's thesis, Massachusetts Institute of Technology, 1988.Google Scholar
  4. [4]
    J. J. Little, H. H. Bülthoff, and T. Poggio. Parallel optical flow using local voting. In Proc. Int. Conf. on Comp. Vision, pages 454–459, Tarpon Springs, Florida, Dec. 1988. IEEE, Washington, DC.Google Scholar
  5. [5]
    J. J. Little and W. E. Gillett. Direct evidence of occlusion in stereo and motion. Tr-90-5, UBC Dept. of Computer Science, Vancouver, BC, 1990.Google Scholar
  6. [6]
    T. Poggio, E. B. Gamble, and J. J. Little. Parallel integration of vision modules. Science, 242(4877):436–440, October 21 1988.Google Scholar
  7. [7]
    T. Poggio and the staff. The MIT Vision Machine. In Proc. Image Under. Work., Cambridge, MA, April 1988. Morgan Kaufmann, San Mateo, CA.Google Scholar
  8. [8]
    S. B. Pollard, J. E. W. Mayhew, and J. P. Frisby. Disparity gradients and stereo correspondences. Perception, 1987.Google Scholar
  9. [9]
    A. Spoerri and S. Ullman. The early detection of motion boundaries. In Proc. Int. Conf. on Comp. Vision, pages 209–218, London, England, June 1987. IEEE, Washington, DC.Google Scholar
  10. [10]
    J. Weng, N. Ahuja, and T. S. Huang. Two-view matching. In Proc. Int. Conf. on Comp. Vision, pages 64–73, Tarpon Springs, Florida, Dec. 1988. IEEE, Washington, DC.Google Scholar
  11. [11]
    A. L. Yuille and T. Poggio. A generalized ordering constraint for stereo correspondence. A.I. Memo No. 777, Art. Intell. Lab, MIT, 1984.Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1990

Authors and Affiliations

  • James J. Little
    • 1
  • Walter E. Gillett
    • 2
  1. 1.University of British ColumbiaVancouverCanada
  2. 2.MITCambridgeUSA

Personalised recommendations